Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method comprising: identifying, by one or more processor, a current task; obtaining, by the one or more processor, an indicator of a commencement of a switching event, wherein the switching event comprises a transition originating from the current task and concluding at a new task; obtaining, by the one or more processor, behavior analysis data relating to a plurality of past switching events, wherein each past switching event of the plurality of past switching events comprises a transition originating from the current task and concluding at one of a plurality of target tasks, the behavior analysis data comprising a timestamp for each past switching event; determining, by the one or more processor, based on the behavior analysis data, at least one recommended task, wherein the at least one recommended task comprises at least one target task of the plurality of target tasks; and assigning, by the one or more processor, a weighted value to each past switching event based on the timestamps; wherein magnitude of the weighted value assigned to each past switching event corresponds to proximity of the timestamp for each past switching event to a current time; wherein the behavior analysis data further comprises the weighted value of each past switching event; and wherein the determining is performed so that a second of the plurality of past switching events occurring subsequent to a first of the plurality of past switching events is assigned a larger weight than the first of the plurality of past switching events based on second of the plurality of past switching events occurring closer in time to the current time than the first of the plurality of past switching events.
The system monitors a user's task switching. When a task switch starts, the system analyzes past task switches originating from the current task to recommend a new task. This analysis uses timestamps of past switches, giving more weight to recent switches. The system identifies the current task, detects the beginning of a switch to a new task, and retrieves data on past switches from the current task to various target tasks, including timestamps for each past switch. It then calculates a weighted value for each past switch based on how recent it was, with more recent switches having higher weights. Based on these weighted values, it suggests one or more recommended tasks.
2. The method of claim 1 , further comprising: providing, by the one or more processor, the at least one recommended task to a client; obtaining, by the one or more processor, an indicator of a conclusion of the switching event; identifying, by the one or more processor, the new task; and updating, by the one or more processor, the behavior analysis data with information related to the switching event, wherein the information related to the switching event comprises a timestamp of a time of the transition originating from the current task and concluding at the new task.
Building upon the previous task switching system, this enhancement provides the recommended task suggestions to a client (e.g., a user interface). Once the task switch completes, the system detects the end of the switch, identifies the new task, and updates the historical switching data. This update includes adding a timestamp for the most recent switch, recording the time of the transition from the current task to the new task. This ensures the task recommendation system continuously learns and adapts to the user's switching patterns by incorporating the latest task transition data.
3. The method of claim 1 , wherein the at least one recommended task comprises the new task.
In the described task switching system, one of the recommended tasks is the new task that the user is actually switching to. This means the system is suggesting the task the user is already transitioning to, potentially as a confirmation or to provide related tools or information for that task. The system analyzes past task switches originating from the current task to recommend a new task. This analysis uses timestamps of past switches, giving more weight to recent switches.
4. The method of claim 1 , wherein the determining further comprises: selecting, by the one or more processor, the at least one recommended task based on a count of each past switching event of the plurality of past switching events concluding at the at least one recommended task and the weighted value assigned to each past switching event of the plurality of past switching events concluding at the at least one recommended task.
To determine which tasks to recommend in the task switching system, the system considers both the frequency of past switches to a particular task and the recency of those switches. It counts how many times the user has switched to each potential target task from the current task. It also uses the weighted value assigned to each past switch (based on its timestamp) to favor more recent transitions. The system then selects the recommended task based on a combination of the number of times the user has switched to that task in the past and how recently those switches occurred.
5. The method of claim 1 , wherein the timestamp for each past switching event of the plurality of past switching events is within a predefined window of time.
The task switching system only considers past task switches that occurred within a specific time window. For example, only the switches that happened in the last week or the last month are used to make recommendations. This ensures the system focuses on the user's more recent behavior and adapts to changes in their task switching patterns. The system analyzes past task switches originating from the current task to recommend a new task. This analysis uses timestamps of past switches, giving more weight to recent switches.
6. The method of claim 1 , further comprising: determining, by the one or more processor, a category for the current task and for each of the plurality of target tasks, wherein the behavior analysis data further comprises the category for the current task and for each of the plurality of target tasks in the behavior analysis data.
The task switching system categorizes tasks to improve recommendations. It determines a category for the current task and for each potential target task. The historical task switching data also includes the categories for the current task and each of the target tasks. This allows the system to identify patterns of switching between tasks within the same category or between different categories. The system analyzes past task switches originating from the current task to recommend a new task. This analysis uses timestamps of past switches, giving more weight to recent switches.
7. The method of claim 6 , wherein a task of the at least one recommended task and the current task have a common category.
In the category-aware task switching system, the system prioritizes recommending tasks that belong to the same category as the current task. For instance, if the user is currently working on a document, the system might recommend another document or a task related to document editing. The system determines a category for the current task and for each potential target task, wherein the behavior analysis data further comprises the category for the current task and for each of the plurality of target tasks in the behavior analysis data.
8. The method of claim 1 , wherein the timestamp for each past switching event is within a given window of time, and wherein the obtaining behavior analysis data comprises obtaining a portion of data smaller than a predefined maximum amount of data and greater than a predefined minimum amount of data.
The task switching system limits the amount of historical data used to make recommendations. The timestamps for past events must be within a defined window of time. The system obtains a portion of data smaller than a predefined maximum amount of data and greater than a predefined minimum amount of data to ensure only the most relevant data points are considered while avoiding over-reliance on limited data. The system analyzes past task switches originating from the current task to recommend a new task. This analysis uses timestamps of past switches, giving more weight to recent switches.
9. The method of claim 1 , wherein the obtaining an indicator occurs prior to conclusion of the transition originating from the current task and concluding at a new task.
The task switching system starts analyzing and recommending tasks as soon as it detects the beginning of a task switch, even before the user has fully transitioned to the new task. This allows the system to proactively suggest relevant options or tools to facilitate the switch. The system identifies the current task, obtains data on past switches from the current task to various target tasks, and calculates a weighted value for each past switch based on how recent it was, with more recent switches having higher weights. Based on these weighted values, it suggests one or more recommended tasks.
10. The method of claim 1 , wherein the determining occurs prior to conclusion of the transition originating from the current task and concluding at a new task.
The task switching system determines the recommended task before the user completes the transition to a new task. This allows the system to proactively suggest tasks and relevant options before the user has fully committed to a specific new task. The system identifies the current task, obtains data on past switches from the current task to various target tasks, and calculates a weighted value for each past switch based on how recent it was, with more recent switches having higher weights. Based on these weighted values, it suggests one or more recommended tasks.
11. The method of claim 1 , wherein the at least one recommended task is a task for initiation by a user to improve efficiency with which the user switches to the new task.
The task switching system's recommended tasks are aimed at improving the user's efficiency in switching to the new task. The suggestions are designed to help the user quickly and smoothly transition to the new task by providing relevant information, tools, or options. The system analyzes past task switches originating from the current task to recommend a new task. This analysis uses timestamps of past switches, giving more weight to recent switches. The recommended task is a task for initiation by a user to improve efficiency with which the user switches to the new task.
12. The method of claim 1 , wherein the method includes presenting the one or more recommended task as a displayed option on a display for selection by a user, and wherein the method includes performing a recommended task of the one or more recommended task presented on the display responsively to action of the user.
The task switching system presents the recommended tasks as selectable options on a display. The user can then choose one of these options to initiate the corresponding task. When the user selects a recommended task from the display, the system automatically performs that task. The system analyzes past task switches originating from the current task to recommend a new task. This analysis uses timestamps of past switches, giving more weight to recent switches.
13. The method of claim 1 , wherein the determining includes predicting based on the behavior analysis data, for each target task of the plurality of target tasks, a likelihood of the target task being a next task transitioned to by a user, and specifying the one or more recommended task based on the predicting.
This invention relates to task recommendation systems that analyze user behavior to predict and suggest next tasks. The problem addressed is the need for systems that can intelligently recommend tasks to users based on their historical behavior, improving efficiency and user experience. The method involves collecting behavior analysis data from a user's interactions with a system, which may include task completion history, time spent on tasks, and transition patterns between tasks. This data is used to predict the likelihood of a user transitioning to a specific target task from their current task. The system then recommends one or more tasks to the user based on these predictions, prioritizing tasks with higher likelihoods. The recommendation process may also consider additional factors such as task dependencies, user preferences, or system constraints. By dynamically analyzing behavior patterns, the system provides personalized and context-aware task recommendations, reducing decision fatigue and enhancing productivity. The method ensures that recommendations are relevant and aligned with the user's workflow, improving task completion efficiency.
14. The method of claim 1 , wherein the method includes specifying a target task of the plurality of target task as a recommend task of the at least one recommended task based on a predicting that the target task is a likely next task to be transitioned to by a user.
The task switching system predicts the likelihood of the user transitioning to each target task. If the system predicts that a specific target task is likely to be the next task, it will recommend that task. The system analyzes past task switches originating from the current task to recommend a new task. This analysis uses timestamps of past switches, giving more weight to recent switches. The method includes specifying a target task of the plurality of target task as a recommend task of the at least one recommended task based on a predicting that the target task is a likely next task to be transitioned to by a user.
15. A computer program product comprising: a computer readable storage medium readable by one or more processor and storing instructions for execution by the one or more processor for performing a method comprising: identifying, by one or more processor, a current task; obtaining, by the one or more processor, an indicator of a commencement of a switching event, wherein the switching event comprises a transition originating from the current task and concluding at a new task; obtaining, by the one or more processor, behavior analysis data relating to a plurality of past switching events, wherein each past switching event of the plurality of past switching events comprises a transition originating from the current task and concluding at one of a plurality of target tasks, the behavior analysis data comprising a timestamp for each past switching event; determining, by the one or more processor, based on the behavior analysis data, at least one recommended task, wherein the at least one recommended task comprises at least one target task of the plurality of target tasks; and assigning, by the one or more processor, a weighted value to each past switching event based on the timestamps; wherein magnitude of the weighted value assigned to each past switching event corresponds to proximity of the timestamp for each past switching event to a current time; wherein the behavior analysis data further comprises the weighted value of each past switching event; and wherein the determining is performed so that a second of the plurality of past switching events occurring subsequent to a first of the plurality of past switching events is assigned a larger weight than the first of the plurality of past switching events based on second of the plurality of past switching events occurring closer in time to the current time than the first of the plurality of past switching events.
A software program product analyzes a user's task switching to recommend new tasks. When a task switch starts, the program analyzes past task switches originating from the current task, using timestamps to weight recent switches more heavily. The program identifies the current task, detects the start of a switch, retrieves data on past switches from the current task to various target tasks (including timestamps), and calculates a weighted value based on recency. Based on these weights, it suggests recommended tasks. This program is stored on a computer-readable medium and executed by a processor.
16. The computer program product of claim 15 , further comprising: providing, by the one or more processor, the at least one recommended task to a client; obtaining, by the one or more processor, an indicator of a conclusion of the switching event; identifying, by the one or more processor, the new task; and updating, by the one or more processor, the behavior analysis data with information related to the switching event, wherein the information related to the switching event comprises a timestamp of a time of the transition originating from the current task and concluding at the new task.
The software program from the previous description provides task recommendations to a client (like a user interface). After the task switch, it detects the completion, identifies the new task, and updates the historical data with a timestamp of the switch. This constant learning adapts to user patterns. This builds on the program that analyzes task switches to recommend new tasks. When a task switch starts, the program analyzes past task switches originating from the current task, using timestamps to weight recent switches more heavily.
17. The computer program product of claim 15 , wherein the determining further comprises: assigning, by the one or more processor, a weighted value to each past switching event based on the timestamps, wherein magnitude of the weighted value assigned to each past switching event corresponds to proximity of the timestamp for each past switching event to a current time, and wherein the behavior analysis data further comprises the weighted value of each past switching event.
The software program determines the weighted value for past switching events based on the timestamps, and the behavior analysis data further comprises the weighted value of each past switching event. When a task switch starts, the program analyzes past task switches originating from the current task, using timestamps to weight recent switches more heavily. The program identifies the current task, detects the start of a switch, retrieves data on past switches from the current task to various target tasks (including timestamps), and calculates a weighted value based on recency. Based on these weights, it suggests recommended tasks. This program is stored on a computer-readable medium and executed by a processor.
18. A system comprising: a memory; one or more processor in communication with the memory; and program instructions executable by the one or more processor via the memory to perform a method, the method comprising: identifying, by the one or more processor, a current task; obtaining, by the one or more processor, an indicator of a commencement of a switching event, wherein the switching event comprises a transition originating from the current task and concluding at a new task; obtaining, by the one or more processor, behavior analysis data relating to a plurality of past switching events, wherein each past switching event of the plurality of past switching events comprises a transition originating from the current task and concluding at one of a plurality of target tasks, the behavior analysis data comprising a timestamp for each past switching event; determining, by the one or more processor, based on the behavior analysis data, a recommended task, wherein the recommended task comprises a target task of the plurality of target tasks, and wherein the determining includes predicting, for each target task of the plurality of target tasks a likelihood of the target task being a next task transitioned to, and specifying as the recommended task a target task of the plurality of target tasks predicted most likely to be the next task; and assigning, by the one or more processor, a weighted value to each past switching event based on the timestamps; wherein magnitude of the weighted value assigned to each past switching event corresponds to proximity of the timestamp for each past switching event to a current time; wherein the behavior analysis data further comprises the weighted value of each past switching event; and wherein the determining is performed so that a second of the plurality of past switching events occurring subsequent to a first of the plurality of past switching events is assigned a larger weight than the first of the plurality of past switching events based on second of the plurality of past switching events occurring closer in time to the current time than the first of the plurality of past switching events.
A computer system with memory and a processor predicts task switches and recommends tasks. The system identifies the current task and detects the start of a switch. It analyzes past switches from the current task to target tasks, using timestamps for each switch. It then predicts the likelihood of each target task being the next task and recommends the most likely one. The system weighs past switches based on recency, giving more weight to recent ones. The likelihood prediction informs the recommendation. The method comprises assigning, by the one or more processor, a weighted value to each past switching event based on the timestamps.
Unknown
October 10, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.